Preliminary analysis of RGB-NIR Image Registration techniques for off-road forestry environments
This work addresses image registration for sensor-fusion in off-road forestry autonomy, but it is incremental as it primarily evaluates existing methods without introducing new approaches.
The paper evaluated classical and deep learning-based RGB-NIR image registration techniques for off-road forestry environments, finding that methods like NeMAR and MURF showed partial success but faced challenges such as GAN loss instability and difficulties with fine details in dense vegetation.
RGB-NIR image registration plays an important role in sensor-fusion, image enhancement and off-road autonomy. In this work, we evaluate both classical and Deep Learning (DL) based image registration techniques to access their suitability for off-road forestry applications. NeMAR, trained under 6 different configurations, demonstrates partial success however, its GAN loss instability suggests challenges in preserving geometric consistency. MURF, when tested on off-road forestry data shows promising large scale feature alignment during shared information extraction but struggles with fine details in dense vegetation. Even though this is just a preliminary evaluation, our study necessitates further refinements for robust, multi-scale registration for off-road forest applications.